Hi,
I have data with 400k rows and 1200 columns. I developed models with Xgboost and Catboost. Xgboost needs 2-3 hours to run, while Catboost finished training during 6 minutes.
Why Xgboost requires such a big time for training compare to Catboost?
Based on this comparison in Python, I'd expect Catboost to be faster but not by that much. It's hard to say why due to not knowing how the two models were run.